Accurate prediction of taxiing time is important in ensuring efficient and safe operations on the airport surface. It helps improve ground operation efficiency, reduce fuel waste, and improve carbon emissions at the airport. In actual operations, taxiing time is influenced by various factors, including a large number of categorical features. However, few previous studies have focused on selecting such features. Additionally, traditional taxiing time prediction methods are often black-box models that only provide a single prediction result; they fail to provide effective practical references for controllers. Therefore, this paper analyses the features that affect taxiing time from different data types and forms a taxi feature set consisting of nine key features. We also propose a taxiing time prediction method based on adaptive scenario matching rules. This process classifies the scenarios into multiple typical historical scenario sets and adaptively matches the current target scenario to a typical scenario set based on quantified rules. Then, based on the matching results, a pre-trained model obtained from the corresponding scenario set is used to predict the taxiing time of an aircraft in the target scenario, aiming to mitigate the impact of data heterogeneity on prediction results. Experimental results show that compared to baseline methods, the mean absolute error and root mean square error of the proposed method decreased by 4.8% and 12.6%, respectively. This method significantly reduces the fluctuations in results caused by sample heterogeneity and enhances controllers’ acceptance of prediction results from the model. It can be used to further improve auxiliary decision making systems and enhance the precise control capabilities of airport surface operations.

With the continuous increase in airport operational pressure, frequent conflicts and declining efficiency in aircraft surface taxiing have become increasingly prominent. This paper addresses the problem of aircraft conflict resolution and taxi trajectory optimization in airport surface operations by constructing a surface trajectory optimization model that incorporates airline fairness and designing a corresponding Markov decision process. On this basis, an asynchronous guided hierarchical multi-agent reinforcement learning method is proposed to achieve the joint optimization of conflict-priority determination and aircraft taxi trajectory planning through upper- and lower-level coordinated decision-making. Simulation results under the selected peak-hour scenario show that, compared with the first-come, first-served (FCFS) method, the proposed approach improves the departure sequence, waiting time, and number of turns by 40%, 23.61%, and 7.52%, respectively, while also taking into account the interests of different airlines. Further sensitivity analysis indicates that, under disturbances such as increased flight volume and random stand changes, the model still maintains a certain degree of optimization capability and robustness within the same simulation framework, although the optimization performance declines as the disturbance level increases. Algorithm comparison and ablation experiments show that the proposed method has advantages over common baseline methods in convergence speed, training stability, and final optimization performance under the reported experimental settings. The results provide preliminary evidence for the feasibility of introducing fairness constraints into airport surface trajectory optimization under peak-hour operating conditions and offer a useful reference for airport surface conflict resolution and operational optimization.